In general, most of the existing convolutional neural network (CNN)-based deep-learning models suffer from spatial-information loss and inadequate feature-representation issues. This is due to their inability to capture multiscale-context information and the exclusion of semantic information throughout the pooling operations. In the early layers of a CNN, the network encodes simple semantic representations, such as edges and corners, while, in the latter part of the CNN, the network encodes more complex semantic features, such as complex geometric shapes. Theoretically, it is better for a CNN to extract features from different levels of semantic representation because tasks such as classification and segmentation work better when both simple and complex feature maps are utilized. Hence, it is also crucial to embed multiscale capability throughout the network so that the various scales of the features can be optimally captured to represent the intended task. Multiscale representation enables the network to fuse low-level and high-level features from a restricted receptive field to enhance the deep-model performance. The main novelty of this review is the comprehensive novel taxonomy of multiscale-deep-learning methods, which includes details of several architectures and their strengths that have been implemented in the existing works. Predominantly, multiscale approaches in deep-learning networks can be classed into two categories: multiscale feature learning and multiscale feature fusion. Multiscale feature learning refers to the method of deriving feature maps by examining kernels over several sizes to collect a larger range of relevant features and predict the input images’ spatial mapping. Multiscale feature fusion uses features with different resolutions to find patterns over short and long distances, without a deep network. Additionally, several examples of the techniques are also discussed according to their applications in satellite imagery, medical imaging, agriculture, and industrial and manufacturing systems.

Robot-assisted surgery is gaining popularity worldwide and there is increasing scientific interest to explore the potential of soft continuum robots for minimally invasive surgery. However, the remote control of soft robots is much more challenging compared with their rigid counterparts. Accurate modeling of manipulator dynamics is vital to remotely control the diverse movement configurations and is particularly important for safe interaction with the operating environment. However, current dynamic models applied to soft manipulator systems are simplistic and empirical, which restricts the full potential of the new soft robots technology. Therefore, this article provides a new insight into the development of a nonlinear dynamic model for a soft continuum manipulator based on a material model. The continuum manipulator used in this study is treated as a composite material and a modified nonlinear Kelvin–Voigt material model is utilized to embody the visco-hyperelastic dynamics of soft silicone. The Lagrangian approach is applied to derive the equation of motion of the manipulator. Simulation and experimental results prove that this material modeling approach sufficiently captures the nonlinear time- and rate-dependent behavior of a soft manipulator. Material model-based closed-loop trajectory control was implemented to further validate the feasibility of the derived model and increase the performance of the overall system.

Accurate segmentation of pulmonary nodule within medical imagery is of great significance for classification and diagnosis. This task is profoundly challenging due to scarcity of annotated data, the diversity of lung nodule features, and the complexity of the background, making it very difficult to accurately and automatically segment lung nodules. Fortunately, deep learning is widely used in the field of pulmonary nodule segmentation and has achieved good and promising results. In this study, we conducted a comprehensive review of pulmonary nodule segmentation in scholarly journals and conferences within the domains of engineering and computer science over the past five years. Our focus was on the application of various deep-learning architectures for pulmonary nodule segmentation and the exploration of diverse deep learning models tailored to the unique challenges posed by this task. We summarized the pivotal methodologies underpinning lung nodule segmentation based on deep learning, including data preprocessing, network architecture, and loss function. Additionally, we discussed the advantages and disadvantages of different models. Finally, we identified research gaps and outlined future directions in the field of deep learning-driven pulmonary nodule image segmentation, aiming to provide researchers with a comprehensive evaluation of recent studies.

The first demonstration of a pneumatic soft continuum robot is integrated in series with a rigid robot arm, safely performing teleoperated diathermic tissue-cutting. The rigid arm autonomously maintains a safe tool contact force, while the soft arm manually follows the desired cutting path. Ex-vivo experimentation demonstrates submillimetric deviations from target paths.

The main objective of this study is to develop a shared design of master and slave devices for bilateral teleoperation mechanisms used for robotic excavators in agricultural applications. In robot teleoperation research, many potential applications within controlled and hazardous environments come to light. Robots, with their capacity for remote control by human operators through master devices, often employ the master-slave teleoperation approach. This strategy finds frequent use across manufacturing, construction, and agriculture industries. The master-slave system necessitates two interdependent components that collaboratively steer the robot in real time. However, challenges arise when manipulating the robot arm proves challenging due to structural differences between the master device (such as a joystick) and the slave device (the robot arm). A stable operational framework must be established for the robot to function optimally. This teleoperation system employs a master device to govern the actions of the slave device, a dynamic that heavily influences the operational complexity. Hence, the focal point of this study is to enhance the master-slave algorithm for teleoperation applications that rely on controlling robot arm movements. Despite the differing dimensions of the master and slave devices, they both share a common structure. The kinematic model bridging these components must be intelligible to ensure user-friendliness, facilitating effortless robot control. Calculating the robot arm's end effector movement and positioning involves employing the forward kinematics of the arm, determined through Denavit-Hartenberg parameters and transformation matrices. By mitigating communication delays between the master and slave devices using a technique centered around the robot arm's end effector position, the effectiveness of teleoperation can be significantly improved. Our designed robot arm attains 80% to 100% precision across joints. In summary, streamlining the robot arm's structure and minimizing delays offers a route to bolstering both stability and efficiency in robotic movement.